Convergence of Distributed Learning Algorithms for Optimal Wireless Channel Allocation

In this paper we establish the convergence to an optimal non-interfering channel allocation of a class of distributed stochastic algorithms. We illustrate the application of this result via (i) a communication-free distributed learning strategy for wireless channel allocation and (ii) a distributed learning strategy that can opportunistically exploit communication between nodes to improve convergence speed while retaining guaranteed convergence in the absence of communication